The major aim of this proposal is to investigate the interactions between two brain regions involved in fearful and anxious behaviors. The bed nucleus of the stria terminalis (BNST) has been implicated in processing adaptive and pathological anxiety. The central nucleus of the amygdala (CE) contributes to both anxious behaviors and fear learning. These two structures have similar cell types, are interconnected and share almost identical patterns of efferent targets. Despite their anatomical and functional similarities, the interaction between these two structures is poorly understood. We propose to examine the BNST-CE pathway using multiple complementary methods. We will characterize the neurons in the oval nucleus of the BNST that project to the CE by retrogradely labeling BNST neurons and using immunofluorescent analysis of neuropeptide markers. We will also characterize neurons in the CE which receive input from BNST neurons. We will then use optogenetic techniques to specifically activate those neurons in the oval BNST which connect with the CE. An AAV vector expressing Cre recombinase fused to a transcellular tracer protein will be infused into the CE. Cre-recombinase would be transported along with the tracer to cell bodies in the BNST as well as other areas which project to the CE. Second, a Cre-dependent virus expressing the rhodopsin derivative channel-rhodopsin will be placed in the oval nucleus of the BNST. Thus, opsin expression will be Cre-dependent and will be limited to those neurons in the BNST which are anatomically connected to the CE. Success of this dual viral strategy will be verified using immunofluorescence and in vitro electrophysiology. Finally, we will examine behaviorally how the BNST-CE pathway modulates anxiety-like behavior, fear learning, fear generalization and fear extinction using optogenetic techniques to activate this pathway. Together, the experiments in this proposal represent an important step in understanding the functional significance of the BNST-CE pathway. Many anxiety disorders in humans can be characterized by abnormalities in acquiring and/or extinguishing conditioned fear responses. Thus, understanding how anxiety and fear learning interact at the neuronal level has clear translational significance.